Quantitative Measurement of Perineural Invasion for Prognosis Analysis of Oral Cavity Cancer Treated by Radical Surgery with or without Adjuvant Therapy

Abstract

<h3>Purpose/Objective(s)</h3> We aimed to evaluate a quantitative analysis of perineural invasion of cancer cells (PNI) in specimen from radical surgery and prognosis of oral squamous cell carcinoma (OSCC). <h3>Materials/Methods</h3> Cancer registry data were reviewed from 2009 – 2015. Inclusion criteria included oral cavity cancer, treated by radical surgery, PNI was noted, and pathologic sample were available for S100 staining. Patients with M1 disease, synchronous or metachronous cancer were excluded. All pathologic samples were reviewed to confirmed PNI status and processed with immunohistochemistry staining of S100 protein to measure the amount of PNI. All pathologic information and staging results were also reviewed, and clinical outcome were analyzed. <h3>Results</h3> Overall, 92 patients were included. Among them, intra-tumoral PNI (IPNI) and extra-tumoral PNI (EPNI) were found in 63 patients and 29 patients, respectively. The average S100 spots were higher in EPNI group than IPNI group (6.7 vs 3.8, t-test 2-tail significance = 0.021). The 5-year overall survival (OS) and recurrence-free survival (RFS) rates of all patients were 77% and 74.2%. Univariate analysis showed that pathological T4 or N2 stage correlated with poor OS, while only 4 or more spots of S100 per slide correlated with poor RFS. In multivariate analysis, clinical T4 or N2 stage were still correlated with poor OS significantly, while average PNI per slide (APNI) ≥ 4 were the only independent factor of poor RFS. The 5-year RFS were 83.1% and 61.7% for disease with APNI < 4 or ≥ 4, respectively (p = 0.008). <h3>Conclusion</h3> For patients who had OSCC with PNI, more PNI identified by S100 staining indicated poorer RFS, regardless of stage and other prognostic factors. A quantitative PNI identification by S100 IHC result can be used for prognosis prediction. Prospective studies are warranted to verified the prognostic effect of our finding and the solution to improve treatment outcome.